Mobile device locator

Abstract

Examples herein involve estimating a first position of a mobile device based on first communication signals, assigning a first set of particles to a number of respective first sampling locations within a threshold distance of the first position, adjusting the assignment of the first set of particles to second sampling locations based on movement of the mobile device, and estimating a second position of the mobile device based on the second sampling locations.

1 . A method comprising:
 performing an analysis of a movement of a mobile device that includes:   estimating a first position of the mobile device within an indoor setting based on first communication signals by a fingerprint locator, wherein the first communication signals are first Bluetooth Low Energy (BLE) signals detected and provided by a plurality of BLE beacons positioned within the indoor setting;   assigning a first set of particles to a number of respective first sampling locations within a threshold distance of the first position;   applying weights to the first set of particles based on their proximity to the first position of the mobile device estimated by the fingerprint locator;   adjusting the assignment of the first set of particles to second sampling locations based on movement of the mobile device by removing a number of particles having weights that satisfy a first weight threshold and adding a same number of new particles to the second sampling locations within a threshold distance of particles having weights that satisfy a second weight threshold, the adjusting including application of a respective random drift to at least some of the first set of particles, wherein a random drift applied to at least one first particle among the first set of particles is different from a random drift applied to at least one second particle among the first set of particles;   determining to cease the analysis of the movement of the mobile device, the determining further comprising:
 determining that a threshold period of time has elapsed since the mobile device has moved; and 
 determining that a position of the mobile device has been estimated a threshold number of times; and 
   after ceasing the analysis of the movement of the mobile device, calculating an estimated new position of the mobile device based on the application of the random drift and the analysis of the movement of the mobile device.   
     
     
         2 . The method of  claim 1 , further comprising:
 estimating a third position of the mobile device within the indoor setting based on second communication signals, wherein the second communication signals are second BLE signals detected and provided by the plurality of BLE beacons; and   estimating a fourth position of the mobile device based on third sampling locations.   
     
     
         3 . The method of  claim 1 , the second set of particles comprising a particle from the first set of particles, wherein the weight of the particle from the first set of particles satisfies a weight threshold. 
     
     
         4 . The method of  claim 1 , further comprising detecting the movement of the mobile device based on inertial measurements of an accelerometer of the mobile device or inertial measurements of a gyroscope of the mobile device. 
     
     
         5 . The method as defined in  claim 1 , further comprising:
 determining the first sampling locations to randomly assign the first set of particles to the first sampling locations within an area of the first position, the area defined by the threshold distance.   
     
     
         6 . The method as defined in  claim 1 , wherein the first position and the second position are located indoors. 
     
     
         7 . The method of  claim 1 , further comprising estimating the first position by:
 receiving the communication signals from a plurality of beacons;   comparing received signal strength indications of the communication signals to received signal strength indications of a fingerprint database; and   estimating the first position based on location information corresponding to the received signal strength indications in the fingerprint database.   
     
     
         8 . An apparatus comprising:
 a fingerprint locator to estimate a first position of a mobile device based on received signal strengths of communications signals in comparison to received signal strengths of a fingerprint database, wherein the first position is estimated based on a discrete probability distribution of the received signal strengths of communication signals;   a sensor analyzer to measure movement of the mobile device;   a particle allocator to control particles of a particle filter based on the first position of the mobile device and movement of the mobile device, wherein the particle allocator is further to:
 assign weights to the particles, the weights corresponding to distance between the first position and the respective particles; 
 resample the particles by removing a number of particles having weights that satisfy a first weight threshold and adding the same number of new particles to new sampling coordinates within a threshold distance of particles having weights that satisfy a second weight threshold, the resampling including application of a respective random drift to at least some of the particles, wherein a random drift applied to at least one first particle among the particles is different from a random drift applied to at least one second particle among the particles; 
   a movement analyzer to analyze movement of the mobile device until a threshold period of time has elapsed since the mobile device has moved and a position of the mobile device has been estimated a threshold number of times; and   a position calculator to estimate a second position of the mobile device based on sampling coordinates of the particles.   
     
     
         9 . The apparatus of  claim 8 , wherein the position calculator is to estimate the second position by averaging the sampling coordinates, the sampling coordinates corresponding to sampling locations of the particles determined based on the movement of the mobile device. 
     
     
         10 . A non-transitory machine readable storage medium comprising instructions that, when executed, cause a machine to at least:
 estimate a first position of a mobile device by comparing received signal strength indications of communication signals to received signal strength indications of a fingerprint database, wherein the first position is estimated based on a discrete probability distribution of the received signal strength indications of communication signals;   allocate particles to sampling locations within a threshold distance of the first position, the particles comprising respective weights, wherein allocating particles further comprises:
 determining the weights of the respective particles based on the distance between the first position and the respective particles; and 
 removing a portion of the particles based on the weights of the particles of the portion of the particles failing to satisfy a weight threshold; 
   analyze movement of the mobile device based on inertial measurements from sensors of the mobile device;   move the particles to new sampling locations in accordance with the movement of the mobile device, the moving including application of a respective random drift to at least some of the particles, wherein a random drift applied to at least one first particle among the particles is different from a random drift applied to at least one second particle among the particles;   determine to cease analyzing movement of the mobile device, the determining further comprising:
 determine that a threshold period of time has elapsed since the mobile device has moved; and 
 determine that a position of the mobile device has been estimated a threshold number of times; and 
   calculate a new position of the mobile device based on the new sampling locations.   
     
     
         11 . The non-transitory machine readable storage medium of  claim 10 , wherein the instructions, when executed, further cause the machine to allocate the particles to the sampling locations by:
 assigning new particles to replace the portion of the particles to new sampling locations within a proximity of a portion of the particles that satisfy the weight threshold.